Microsoft Technology Licensing, LLC

Kenneth Hiroshi Eguro

Arvind Arasu

Donald Kossmann

Ravishankar Ramamurthy

Raghav Kaushik

Ramarathnam R Venkatesan

Manas Rajendra Joglekar

Panagiotis Antonopoulos

Ravi Ramamurthy

Kaushik Shriraghav

Ray A Bittner, Jr

Shriraghav Kaushik

Shawn Michael Swilley

Spyridon Blanas

Aditya Nigam

Nikhil Vithlani

Pingfan Meng

Vineet Pandey

George E Smith

Prasang Upadhyaya

Douglas Christopher Burger

Eric S Chung

Cesar Alejandro Galindo-Legaria

Roy Peter D'Souza

Alessandro Forin

Blake Douglas Pelton

Haohai Yu

Michael I Zwilling

Joachim Hammer

Bala Neerumalla

Chirag Varde

Jakub J Szymaszek

Zsolt Istvan

Kaan Kara

Kedar Dubhashi

Rehan Ahmed

Ji Sun

Donald Alan Kossmann

Deferred verification of the integrity of data operations over a set of data that is hosted at an untrusted module (UM) is controlled. The controlling includes generating a request for a data operation on the set of data. The request includes an authentication portion. The request is sent to the UM. A response to the request is received from the UM. The response includes cryptographic verification information attesting the integrity of the data operation with respect to prior data operations on the set of data. The response includes results from deferred verification at a trusted module (TM).

Controlling verification of key-value stores

Methods, systems, apparatuses, and computer program products are provided for generating an instruction set for an evaluation engine. An arithmetic expression that combines multiple columns of data (e.g., a first column of data, a second column of data, etc.) is received. Instructions may be generated, that, when executed by an integrated-circuit-based processor, cause the integrated-circuit-based processor to evaluate the arithmetic expression. In examples, a set of instructions may be generated for each column of data represented in the arithmetic expression. For instance, the instructions may comprise a first set of instructions associated with the first column of data, a second set of instructions associated with the second column of data, and so on. The instructions may specify one or more parameters for operations associated with each column of data, such as operations to load data from a buffer, store data into a buffer, arithmetic operations to perform on data, etc.

Column data driven arithmetic expression evaluation

Methods, systems, and computer-readable media are directed towards receiving, at an untrusted component, a query for a data store. The query includes a plurality of data operations. The data store is accessible by the untrusted component. A first proper subset of data operations is determined from the plurality of data operations that do not access sensitive data within the data store. A second proper subset of data operations is determined from the plurality of data operations that access sensitive data within the data store. The first proper subset of data operations is executed, at the untrusted component, to create first results. The second proper subset of data operations is sent to a trusted component for execution. Second results based on the sending the second proper subset of data operations are received from the trusted component. Results to the query are returned based on the first results and the second results.

Secure data processing on sensitive data using trusted hardware

A 'Database Confidentiality System' provides various techniques for using server-side trusted computing in combination with configurable type metadata and user- or system-definable rules associated with individual database fields to implement database confidentiality. In various implementations, type metadata and one or more rules are added to each database field. Metadata includes a domain, method of encryption, and a pointer to an encryption key used to encrypt the data in the corresponding field. The rules define one or more operations allowed on the corresponding data types. The type metadata and rules are optionally integrity protected and/or encrypted to avoid unauthorized changes or access. Various encryption techniques (e.g., probabilistic, Paillier, etc.) allow some computations to be performed in an untrusted environment without access to the encryption key. This enables the Database Confidentiality System to maintain database confidentiality while performing distributed computation and communications between the untrusted machine and the trusted machine.

Type-based database confidentiality using trusted computing

A database management system (DBMS) run a host CPU and a hardware coprocessor accelerate traversal of a tree-type data structure by allocating reusable memory in cache to store portions of the tree-type data structure as the tree-type data structure is being requested by the host CPU. The hardware coprocessor manages the cached tree-type data structure in a manner that is transparent to the host CPU. A driver located at the host CPU or at a separate computing device can provide an interface between the host CPU and the hardware coprocessor, thus reducing communications between the host CPU and the hardware coprocessor.

Systems and methods for providing distributed tree traversal using hardware-based processing

Computer systems, devices, and associated methods of evaluating an expression comprising restricted data are disclosed herein. In one embodiment, a method includes receiving a database statement from a client application and verifying the authenticity of the database statement. If the database statement is authentic, an approved expression is identified in the database statement for creating an evaluation rule. The method further includes restricting evaluation of expressions in a protected computing environment according to the created evaluation rule.

Expression evaluation of database statements for restricted data

Computer systems, devices, and associated methods of optimizing the execution of instructions of a database statement by a database server are disclosed herein. In one embodiment, a method includes identifying a potential execution plan for executing instructions of the database statement and estimating a cost for executing the execution plan. The cost can comprise an encrypted data processing cost associated with a operation in the execution plan of executing an operation on encrypted data in a protected computing environment. The method can include estimating the encrypted data processing cost in the protected computing environment based on statistics generated in the protected computing environment about a database table. In response to estimating the cost for executing the execution plan, comparing the cost to estimated costs of alternative execution plans, selecting the lowest-cost plan for execution, and executing the lowest-cost execution plan.

Execution optimization of database statements involving encrypted data

A number of transmissions of secure data communicated between a secure trusted device and an unsecure untrusted device in a DBMS is controlled. The data is communicated for database transaction processing in the secure trusted device. The number of transmissions may be controlled by receiving, from the untrusted device, an encrypted key value of a key and a representation of an index of a B-tree structure, decrypting, at the trusted device, the key and one or more encrypted index values, and initiating a transmission, a pointer value that identifies a lookup position in the index for the key. The index comprises secure, encrypted index values. Other optimizations for secure processing are also described, including controlling available computation resources on a secure trusted device in a DBMS and controlling transmissions of secure data that is communicated between a secure trusted device and an unsecure untrusted device in a DBMS.

Controlling secure processing of confidential data in untrusted devices

The techniques discussed herein facilitate the transmission, storage, and manipulation of data in an encrypted database management system (EDBMS). An untrusted machine is connected to a data store having encrypted records, a client machine that sends encrypted queries, and a trusted machine that receives and decrypts the encrypted records and encrypted queries. The trusted machine processes the query using semantically secure query operators to produce a query result. The trusted machine ensures the size of the query result conforms to an upper bound on the number or records in the query result and returns the query result.

Customized trusted computer for secure data processing and storage

To protect customer data and provide increased workflow security for processing requested by a customer, a secure communicational channel can be established between a customer and one or more hardware accelerators such that even processes executing on a host computing device hosting such hardware accelerators are excluded from the secure communicational channel. An encrypted bitstream is provided to hardware accelerators and the hardware accelerators obtain therefrom cryptographic information supporting the secure communicational channel with the customer. Such cryptographic information is stored and used exclusively from within the hardware accelerator, rendering it inaccessible to processes executing on a host computing device. The cryptographic information can be a shared secret, an appropriate one of a pair of cryptographic keys, or other like cryptographic information. Similarly, the encrypted bitstream can comprise the cryptographic information, computer-executable instructions executable by the processing circuitry of the hardware accelerator to derive such cryptographic information, or combinations thereof.

Protecting communications with hardware accelerators for increased workflow security

The subject disclosure is directed towards loading parallel memories (e.g., in one or more FPGAs) with multidimensional data in an interleaved manner such that a multidimensional patch/window may be filled with corresponding data in a single parallel read of the memories. Depending on the position of the patch, the data may be rotated horizontally and/or vertically, for example, so that the data in each patch is consistently arranged in the patch regardless of from which memory each piece of data was read. Also described is leveraging dual ported memory for multiple line reads and/or loading one part of a buffer while reading from another.

Parallel memories for multidimensional data access

The subject disclosure is directed towards performing connected components in hardware, such as an FPGA, which is facilitated by a linked list structure that does not grow. During a connected components graph labeling process, when a merge is encountered, the data structure comprising labels and associated equivalency data swaps the equivalency data of the two vertices whose different labels produced the merge condition.

Hardware-amenable connected components labeling

Growing community of inventors

Seattle, WA, United States of America

Kenneth Hiroshi Eguro